The dataset consists of image chips extracted from Planet satellite imagery collected over the San Francisco Bay and San Pedro Bay areas of California. It includes 4000 80x80 RGB images labeled with either a "ship" or "no-ship" classification. Image chips were derived from PlanetScope full-frame visual scene products, which are orthorectified to a 3 meter pixel size.
Here we follow this tutorial
A slightly more sophisticated method, introducing R-CNN image segmentation and object detection to improve ship detection.
Here we follow this tutorial and these notes.